OSJ 2019 : Practical Resilience Index for Coral Reef Assessment
- 1. Available online at http://link.springer.comOcean Sci. J. (2019) http://dx.doi.org/10.1007/s12601-019-0002-1 pISSN 1738-5261 eISSN 2005-7172 Article Practical Resilience Index for Coral Reef Assessment Imam Bachtiar1,2 *, Suharsono2 , Ario Damar3 , and Neviaty P. Zamani4 1 Department of Math and Science Education, Faculty of Teacher Training and Education, University of Mataram, Kota Mataram 83125, Indonesia 2 Department of Magister Pendidikan IPA, Program Pascasarjana, University of Mataram, Kota Mataram 83125, Indonesia 3 Research Center for Oceanography, Indonesian Institute of Sciences, Jakarta 12710, Indonesia 4 Department of Aquatic Resources Management, Faculty of Fisheries and Marine Science, Bogor Agricultural University, West Java 16680, Indonesia 5 Department of Marine Science and Technology, Faculty of Fisheries and Marine Science, Bogor Agricultural University, West Java 16680, Indonesia Received 14 January 2018; Revised 9 May 2018; Accepted 27 September 2018 KSO, KIOST and Springer 2019 Abstract Assessing coral reef resilience is an increasingly important component of coral reef management. Existing coral reef resilience assessments are not practical, especially for developing countries.South-eastAsiancountrieshavebeenusingline-intercept- transect (LIT) in coral reef monitoring for a long time. The present studyproposesanindexforassessingcoralreefresiliencebasedon data collected from the LIT method. The resilience index formula was modified from an existing resilience index for soil communities developed by Orwin and Wardle. We used an ideal resilient coral reef community as a reference point for the index. The ideal coral reefwasdefinedfromdatacollectedfrom1992to2009.Sixvariables were statistically selected for the resilience indicators: coral functional group (CFG), coral habitat quality (CHQ), sand-silt cover (SSC), coral cover (COC), coral small-size number (CSN), and algae- other-fauna (AOF) cover. Maximum values of five variables were determinedasthebeststate,whilethemaximumvalueofCSNwas determinedfrom1240data-setsofIndonesianreefs.Theresilience index performed well in relation to changes in COC, AOF, and SSC variables. Managers can use this tool to compare coral reef resilience levels among locations and times. This index would be applicable for global coral reef resilience assessment. Keywords recovery, food security, reef health, indicators, monitoring 1. Introduction Coral reef ecosystems are increasingly affected by global climate change (Kleypas et al. 1999; Hoegh-Guldberg et al. 2007) and human-induced stress and disturbances (Jackson 1997;Jacksonetal.2001).Coralreefmanagementcanmitigate the potential impact of these disturbances by improving coralreefresilience(Hughesetal.2007;Hoegh-Guldberget al. 2007) regardless of its ecological limitations (Mora et al. 2016). Ecological resilience in terms of capacity to recover from disturbance to the same level of organization (Holling 1973) is increasingly important in coral reef management. The theory of resilience has improved to the point where it should be able to be implemented in coral reef management practices (Nystrom et al. 2008). Resilience assessment of coral reef ecosystems, however, is still at a development stage (Lam et al. 2017). At present, two methods have been available for measuring coral reef resilience.OburaandGrimsditch(2009)providedacomprehensive method which included 61 variables and 5 protocols. This large number of variables and protocols requires significant financial support and considerable expertise, unlikely to be available in developing countries. Maynard et al. (2010) developed a more practical method in assessing coral reef resilience.Thisonly needsa discussionforum tocollectdata based on the personal judgment ofmanagers, scientists, policy makers, and other important stakeholders. This method could be very practical in many countries, but it needs people with sufficient knowledge in coral reef ecology. In many developing countries where coral reef scientists are rare, the Maynard method might still be hard to implement. McClanahan et al.*Corresponding author. E-mail: imambachtiar@unram.ac.id
- 2. 2 Bachtiar, I. et al. (2012) listed 11 key factors of coral reef resilience, but how thefactorswillbeusedinmanagementisnotyetformulated. Cumming et al. (2016) reviewed resilience theory on social- ecological systems that might be very difficult to use in managementbecauseofitscomplexity.Anewpracticalresilience assessment is therefore needed for coral reef management in developing countries with few coral reef ecologists. The aim of the present study was to formulate an index for assessing the resilience state of a coral reef ecosystem using multivariate data extracted from line intercept transects (LIT). LIT has been widely used and extensively learned about in the region of ASEAN countries, including Indonesia (English et al. 1994; Suharsono 2008). This method would be practical in resilience assessment of the ASEAN countries. Despite the range of multivariate data available from LIT (Marsh et al. 1984; English et al. 1994), the majority of extracted data is merely living coral cover (%). Coral reef managers mostly did not knowhow to integrate multivariate data into a single interpretation of the quality of a coral reef ecosystem. Lam et al. (2017) found that existing coral reef resilience assessment and monitoring programs differ as much as 75%. Here, we show that data collected on an ordinary monitoring program can also be used in resilience assessment. Coral reef resilience is defined as recovery potential to a coral predominated community. When using the resilience index developed in the present study, a coral reef manager will be working with multivariate data but interpreting them in a univariate way. 2. Methods Data collection The present study used monitoring data collected by the Research Center of Oceanography (RCO), Indonesian Institute of Sciences (Bahasa Indonesia, LIPI). All the collected data were taken from LIT methods by trained- and supervised- researchers of the same institution. Data were collected within the period of 1992–1998 and 2009; all 1240 transects were 10 m length. This long period may provide samples of the best and the worst coral reefs resilience levels in Indonesia. In the first period LIT were laid at various depths, from 3 to 15 m; in the second period LIT were at ~5 m depth. All positions of transect sites in 2009 were recorded aboard using GPS (Global Positioning System). Theresilienceindexcalculationisbasedonatransectunit. TheLITdatausedinthepresentstudycouldspatiallyrepresent the coral reefs of Indonesian waters (Fig. 1). Data collected in 2009 encompassed coral reefs of the Indian Ocean (Tapanuli Tengah, Nias, Nias Selatan, Mentawai), Indonesian Seas (Sunda Strait, Karimata Strait, Makassar Strait, Java Sea, Fig. 1. Locations of data collection used in the present study. Locations with green background showed data collected in 1992–1998, light-brown background showed data in 2009
- 3. Practical Resilience Index for Coral Reef Assessment 3 FloresSea,ArafuraSea),South ChinaSea (Natuna,Bintan), and Pacific Ocean (Biak, Raja Ampat). The older data (1992–1998) added locations at the Java Sea (Seribu Islands, Karimun Jawa), Sunda Strait (Bakahauni, Merak) and Raja Ampat.Theolderdataprovidedsomereferencetoearliercoral reef conditions when anthropogenic impacts were smaller and the intervals between natural disturbances (bleaching) were longer. These data were all used to compose the “resilient coral reef” as the reference point in the formulation of the resilience index. Data analysis The coral reef resilience index used in this study was a modificationofthesoilcommunityresilienceindexdeveloped by Orwin and Wardle (2004). The modification included redefining indicator variables and constructing a single reference (control) community. The original formula of the Orwin and Wardle (2004) index was RSt = Resiliece index at t, D0 = difference between before and after disturbance at control community, Dx = differerence between before and after disturbance at impact community. The ecological index modification has two main processes, i.e. determining indicator variables and building the formula. Indicator variables were identified from a literature review and classified into six resilience components: biological legacy, structural legacy, and mobile link (Lundberg and Moberg 2003), coral reef regimes (productivity), herbivory, and water quality. These variables are briefly discussed in this paper and listed in Table 1. (a) Biological legacy, i.e. surviving coral colonies. It has been long understood that coral species exhibit different resistance to the same disturbance (Brown and Suharsono 1990; Ninio and Meekan 2002). Surviving colonies could enhance recolonization of opened spaces from both larvae produced (Miller and Mundy 2003; Starger et al. 2010) and vegetative propagation (Golbuu et al. 2007; Williams et al. 2008). Biological legacy is therefore related to biological diversity (Peterson et al. 1998) and larval dispersal. In the present study, diversity at genera level (coral genera richness, CGR) and functional level (coral functional group, CFG) were used to represent biological diversity. Species richness is hard to determine on LIT since coral taxonomists are rare in many developing countries (Erdinger and Risk RSt 2 D0 D0 Dx + ------------------------------- 1–= Table 1. Indicator variables used in development of the coral reef resilience index Components Variables Unit Justifications References a) Biological legacy (Biodiversity) Coral genera richness (CGR) # genera High resistance, surviving colonies contribute to sexual and asexual recruitment Miller and Mundy (2003); Williams et al. (2008); Starger et al. (2010) Coral functional group (CFG) # life form Functional diversity augment robust community Bellwood et al. (2004) b) Structural legacy (Complexity) Coral massive and sub-mas- sive cover (CMS) % CM+CS covers High resistance, high habitat com- plexity, but low water quality Erdinger and Risk (2000); Ninio and Meekan (2002) Unsuitable settlement substrate (SSC) % sand and silt covers Inhibit coral settlement and growth Hodgson (1990); Babcock and Davies (1991) c) Mobile link (Recruitment) Coral colony size (CCS) classes; 10 cm interval #size classes Continuing recruitment increase recovery Peterson et al. (1998) Coral small-size colonies (CSN); D 10 cm #small colonies Existing recruitment enhance recovery Van Moorsel (1985); Golbuu et al. (2007) d) Productivity (Regimes) Coral cover (COC) % coral cover Healthy coral reef, high productivity Done (1992); Hughes (1994) Algal total cover (ATC) % algal cover Unhealthy reef, low herbivory Done(1992);Hughes(1994); Bahartan et al. (2010) Other fauna cover (OTF) % OTF cover Unhealthy reef, low productivity Tkachenko et al. (2007); Cruz et al. (2016) e) Herbivory Macroalgal cover (AMC) % cover Unhealthy reef, low herbivory Hughes et al. (2007); Roff and Mumby (2012) f) Water quality Coral acroporiid cover (CAC) % acroporidae cover Good water quality, rapid recovery, but low habitat complexity Done (1982); Erdinger and Risk (2000); Roff and Mumby (2012)
- 4. 4 Bachtiar, I. et al. 2000). It is assumed that adequate coral larval recruitment will occur with sufficient coral colonies. (b) Structural legacy, i.e. habitat complexity and settlement substrate availability. A complex habitat provides a suitable substrateforlarvalsettlement,hostsfishdiversity(Wilsonet al. 2007), and maintains herbivory processes (Ledlie et al. 2007). Previous studies measured habitat complexity using a spatial index (Rogers et al. 1983), surface index (Robert andOrmond1987),orvisualassessment(Wilsonetal.2007). In thepresentstudy,habitatcomplexitywasindicatedbyabundance of massive- and sub-massive- corals (CMS). Both coral life forms are structurally resistant to biological and physical disturbanceswhenothercorallifeforms(branching,foliose, encrusting, etc.) become rubble. Availability of settlement substrate is also very important to facilitate sexual coral recruitment. Measuring unsuitable settlement substrate (USS) is easier than the opposite, since some suitable settlement substrates may have been temporarily occupied by benthic community. (c) Mobile link, i.e. incoming mobile organism. Planulae larvae are one of the important mobile link organisms that facilitate coral recolonization.Incoming fish and other benthic larvae and juveniles are also important but indirectly related to coral recolonization and therefore do not directly affect coral reef recovery. In the present study, coral colony size was chosen as an indicator of incoming coral recruitment. When recruitment is continuing, the range of the coral colony sizeshouldbewide.Thisconditioncouldbeindicatedbythe number of colony size classes (coral colony size-classes, CCS), at 10 cm intervals. The abundance of the smallest colony size ( 10 cm) could also indicate current coral recruitment. Indicator variable of coral small-size number (CSN) is chosen to represent the mobile link, coral recruitment or colonization. Marsh et al. (1984) provided a formula for measuring colony diameter on a line transect which is too complex for coral reef managers. In this study, colony size was defined as the length of colony size at the intercept. This approach would not affect the results of the study as colony size measurement was not one of the objectives. Confounding data on CSN might happen whether it comes from recruitment or coral fragment. This could also be neglected since index as an ecological indicator will be working on a macro scale instead. McClanahan et al. (2012) also recognized coral size distribution as important factor in coral reef resilience. (d) Intensity of herbivory is an important factor controlling algal abundance on reefs. Using herbivorous fishes as an indicatorvariableofherbivoryintensity,however,mightnot be very appropriate. Phase-shift could happen without reduction on herbivorous fishes abundance, e.g. in the Great Barrier Reefs (Cheal et al. 2010). Important herbivores are dependent on the composition of the algal community (Fox and Bellwood 2008; Hoey and Bellwood 2008). Redundancy of herbivory roles between herbivorous fishes and sea urchins could also vary among locations (Carpenter 1990). A coral reef could still maintain its resilience in conditions of low abundance of herbivorous fishes, when sea urchins are sufficiently abundant (Hughes 1994). Furthermore, important herbivory could be carried out by non-herbivorous fishes, such as Platax pinnatus (Bellwood et al. 2006). In the present study, abundance of macroalgae (AMC) and total algal abundance (ATC) were used as indicator variables of herbivory. Both indicator variables could represent the impact of herbivory intensity and nutrient availability (Littler et al. 2006). (e) Ecosystem productivity. Coral reefs exhibit multi-stable states with different productivity levels. Highly productive coral dominated communities could shift into alternate stable states which are predominated by macroalgae (Hughes 1994), soft corals (Fox et al. 2003), or anemone (Tkachenko et al. 2007). Indicator variables that represent the productivity of a coral reef ecosystem are COC (coral cover), OTF (other fauna cover), and ATC(algae totalcover). A high productivity coral reef should be high in COC but low in OTF and ATC variables. (f) Waterquality.Whiletemperature,chlorophyllandturbidity can be easier to measure than nutrients and contaminants in reef waters, they still require more specialized equipment and training, difficult to access in a developing country. To simplify assessment, a bioindicator can be chosen to represent waterquality.ParticularcoralspeciessuchasAcroporahave been linked with good water quality (Done 1982; Erdinger and Risk 2000). Thus the cover acroporid corals (CAC) were used as an indicator variable of good water quality. Among the 11 variables in Table 1, there were variables thatneedtobecombined.Therewerevariableshavingsmall abundance but high variances, i.e. algae total cover (ATC), macroalgae cover (AMC), and other fauna cover (OTF). These variables are ecologically important but statistically too small to detect their effect. ATC, AMC and OTF variables were therefore combined as one variable AOF (algae and other fauna) so that its importance will not be neglected in statistical analysis.
- 5. Practical Resilience Index for Coral Reef Assessment 5 There were also variables with contrasting characteristics. Coralmassive+sub-massive(CMS)coverandcoralacroporid cover (CAC) play different roles in ecosystem resilience. High CMS cover could indicate high coral reef complexity butitcouldalsoindicatepoorerwaterqualityandslowrecovery. Massive and sub-massive corals found on inshore reefs are relatively tolerant to low water quality from sedimentation and land-based pollution (Erdinger and Risk 2000). In contrast, high CAC could indicate less coral reef complexity but also indicates rapid recovery (Ninio and Meekan 2002). Both CMS and CAC are equally important in determining coral reef resilience. Combining CMS and CAC into a single variable CHQ (coral habitat quality) needs a special approach, based on existing studies on Indo-Pacific reefs. Cumulative coverage of CMS and CAC has a maximum value of 100%. It was therefore assumed that the best resilience of a coral reef would be composed of 50% of CMS and 50% of CAC. This composition ensures rapid coral recovery as CAC (Ninio and Meekan 2002) promotes high growth rates and adaptation (Guest et al. 2016), and CMS promotes habitat complexity by maintaining the abundance and diversity of herbivorous fishes(Ledlieetal.2007;Guestetal.2016).CHQismathematically calculated as follows: CAC=coral Acropora cover, CMC=coral massive cover, CSC=coral sub-massive cover Algal and other fauna (AOF) was defined to integrate AMC,ATCandOTF.DataofIndonesiancoralreefsshowed that each AMC, ATC and OTF variables have a very high coefficientofvariability,i.e.349%,87%,and140%respectively. The three variables were then pooled into a new variable, AOF, which is the sum of total algal cover (ATC) and other fauna cover (OTF). Macroalgal cover (AMC) is part of the ATC. They were separately measured as different variables since macroalgae predominance has been shown to be the most common phase shift condition on coral reefs (Hughes 1994; Hughes et al. 2007). Statistical analysis was applied to refine indicator variables. Pearsoncorrelationindexwascalculatedtoidentifyredundancies among variables. Non-parametric multivariate statistics, BEST (Biological Environmental STepwise) analysis was applied to find the best combinations in variable reduction for several correlated variables. BEST is a multivariate tool to select environmental variables or species that best explain community pattern. Reduction of the number of variables could be carried out using BEST analysis, as long as its correlation coefficient is higher than 95% (Clarke et al. 2008). Prior to data analysis, data were transformed into log (x + 1) and normalized, since some variables had different scales. PCA (Principal Component Analysis) was performed to weight each indicator variable based on its contribution to the first principal component, as suggested in Primpas et al. (2010). All multivariate statistics were carried out using Primer 6 version 6.1.13 software. 3. Results Refining indicators After grouping the five indicator variables (CAC and CMS as CHQ; AMC, ATC, and OTF as AOF), there were 8 remaining indicator variables of coral reef resilience. Among these, three variables were interdependently related, i.e. COC, AOF and SSC. Potentially, coverage of COC and AOF is dependent on the coverage of SSC, i.e. sand and silt covers. Maximum cover of COC and AOF is 100%, when SSC is zero. The maximum cover of COC and AOF decreases with increasing SSC. Among 8 indicator variables of the resilience index, many variables showed significant correlation, as the number of data was robust (N = 1240 transects). Two pairs of variables showed a considerably high correlation index. Correlation indices between CFG and CGR was 0.691, and between COC and CCS was 0.833 (Table 2). The number of variables needs to be reduced, to avoid redundancy and to improve its applicability without significant reduction in the quality of the index. BEST analysis provided combinations of seven and six variables that may be used to reduce the number of variables. The highestpriority for omissionisthe coralgenera richness (CGR) due to difficulties in underwater genera identification. From the BEST combinations (Table 3), while six or seven variables do not reduce the performance very much, in contrast thereisahugedifferenceintheeffortspentonmeasurement. It was decided to omit CGR and CCS from indicators of coral reef resilience. Leaving out CGR and CCS variables the index still maintained representativeness of each resilience component, CFG represents biodiversity and CSN represents mobile link resilience components. CHQ CAC* CMC CSC+ =
- 6. 6 Bachtiar, I. et al. Defining a reference community An ideal ‘resilient’ coral reef must be set up as a reference community of the index. It must show a maximum value of resilience index, which happens when it has maximum values in all six indicator variables. On the other hand, a ‘non-resilient’coralreefmusthaveaminimumvalueofindex. Amongthesixindicatorvariables,themaximumandminimum theoretical values of five variables are known. Coral cover (COC), for example, theoretically has a minimum and maximumvalueof0.00and100.00%respectively(Table4). Coral functional group (CFG), i.e. coral life form, theoretically between 0–13 groups, as it has been standardized in English et al. (1994). The theoretical maximum value of one variable, CSN, however, was not known. The maximum value of CSN was therefore determined from 1240 transects. Since its maximum value from transects was 23 colonies, the maximum value of CSN was set up at 25, which has a very low probability of occurrence (0.048%). Formulating resilience index The coral reef resilience index uses a single reference community, the ideal resilient coral reef community, i.e. a community with the best (maximum) values in all n indicator variables. Orwin-Wardle index modification resulted in a coral reef resilience index as described in Eq. 1. (1) RIj = Resilience index of transect j, Ximax = maximum value of variable Xi., Xi.min = minimum value of variable Xi., RIj 2 Ximax Ximin– Ximax Ximin– Xij Ximin– + ------------------------------------------------------------------------------- 1– i 1= n = Table 2. Matrix of correlation coefficients among eight resilience variables (N = 1060) CGR CFG SSC CHQ CCS CSN COC AOF CGR 1.000 CFG 0.691 1.000 SSC -0.224 -0.277 1.000 CHQ 0.299 0.550 -0.222 1.000 CCS 0.343 0.382 -0.287 0.395 1.000 CSN 0.512 0.443 -0.155 0.055 0.015 1.000 COC 0.511 0.526 -0.389 0.478 0.833 0.175 1.000 AOF 0.046 0.023 -0.113 -0.100 -0.280 0.123 -0.313 1.000 Table 3. Combinations of seven and six variables provided by BEST analysis No. Number variables Correlation index Variable combinations Omitted variable 1 7 0.992 CGR, SSC, CHQ, CCS, CSN, COC, AOF CFG 2 7 0.991 CFG, SSC, CHQ, CCS, CSN, COC, AOF CGR 3 7 0.990 CGR, CFG, SSC, CHQ, CCS, CSN, AOF COC 4 7 0.988 CGR, CFG, SSC, CHQ, CSN, COC, AOF CCS 5 6 0.974 CGR, SSC, CHQ, CSN, COC, AOF CFG, CCS 6 6 0.973 CGR, SSC, CHQ, CCS, CSN, AOF CFG, COC 7 6 0.973 CFG, SSC, CHQ, CSN, COC, AOF CGR, CCS 8 6 0.971 CFG, SSC, CHQ, CCS, CSN, AOF CGR, COC 9 6 0.970 CHQ, CCS, CSN, SSC, COC, AOF CGR, CFG 10 7 0.969 CGR, CFG, CHQ, CCS, SSC, COC, AOF CSN Table 4. Variable composition of the “super coral reef” as the reference point of the resilience index assessment Resilience indicator From transects Unit Minimum Maximum 1) Coral Functional Group (CFG) 0–10 Group 0 13 2) Sand and Silt cover (SSC) 100–0 % 100 0 3) Coral Habitat Quality (CHQ) 0–42 % 0 50 4) Coral Small-size Number (CSN) 0–23 Colony 0 25 5) Coral Cover (COC) 0–100 % 0 100 6) Algae-Other-Fauna cover (AOF) 100–0 % 100 0
- 7. Practical Resilience Index for Coral Reef Assessment 7 Xij = real value of Xi at transect j. Using data of the resilient coral reef, the formula for the coral reef resilience index is defined as follows. + (2) Each indicator variable in Eq. 2 might have a different magnitude of contribution to the index, as they could have different variances. Indicator variables need to be weighted toimproveindexsensitivity.PCAwasperformedtoobjectively weight each of the resilience indicators. PCA showed that the PC1 contributed 47.7% of the total variances. This number determined its liability to be used as a weighting factor. Variables were weighted using PC1 as listed in Table 5. Weighting of indicator variables changed the index value. The maximum value of the resilience index was 1.930, while the minimum was -0.200. A correction factor is required to ensure that the minimum value was close to 0.000, i.e. when a coral reef was 100% covered by SSC (sand and silt). The finalformulaofthecoralreefresilienceindexisasEq.3or4. + + (3) + + (4) Using the present index, a single transect has a theoretically resilienceindexbetween0.000and2.130,butwehardlyfind coralreefwitharesilienceindexmorethan1.00.Assessment of ~1240 transects from Indonesian reefs revealed a range between 0.021 and 1.070, with an average (± SD) of 0.468 ± 0.225, and a median of 0.469. Proportion of transects with a resilience index 1.000 was only 0.403%. This proportion is very small and not usually found in natural habitats. Furthermore, resilience index assessment on a reef must be carried out using more than one transect, as such assessment needs replications. From the present data, at any location (site) with a minimum of three transects, there was no reef with an average resilience index of 1.000. The highest average resilience index (± SD) found was 0.976 ± 0.107. Theoretically, we rarely encounter any reefs with a resilience index 1.000. Validating resilience index Comparing resilience index with COC, AOF and SSC showed a positive relationship between the increase of COC and an increase in the resilience index. On the other hand, an increase in AOF and SSC coincided in a decrease in the resilience index. Recovery of a coral dominated community depends on the composition of the coral community and the other two opposing factors, AOF and SSC. This suggests that coral cover alone may not be a good measure of coral reef resilience since it does not necessarily show diversity of coral functional group and coral recruitment. The resilience index provides a more meaningful metric forassessmentofresiliencelevelsthancoralcover,although regression analysis for both variables COC and AOF shows its significant dependency. Resilience index increases with increasing coral cover but decreases with increasing algal- other-fauna and sand-silt covers (Fig. 2). Coral reefs with poor coral cover (< 25%) show a wide range of resilience levels, from 0.021 to 0.663. Coral reefs with algal cover > 50%, however, have a resilience index range between 0.100 and 0.802. This high resilience index was found at a transect in Biak (BIALT11) which has 36.0% COC and 52.4% AOF. Coral reef with the best resilience index (1.000) may still have AOF about 0 and 17%, with coral cover ranging from 71.0 to 89.70%. The present resilience index is also readily applicable for extreme situations. An extreme reef which is covered by 100% sand and silt (SSC) will have a resilience index of 0.000. This means that such a reef would never be recolonized RI 2 13 0– 13 0– 13 CFG– + -------------------------------------------------- 1– 2 50 0– 50 0– 50 CHQ– + -------------------------------------------------- 1– += 2 25 0– 25 0– 25 CSN– + ------------------------------------------------- 1– 2 100 0– 100 0– 100 COC– + -------------------------------------------------------- 1– + 2 100 0– 100 0– SSC 0– + ------------------------------------------------ 1– 2 100 0– 100 0– AOF 0– + ------------------------------------------------- 1– – RI 0.56 2 13 0– 13 0– 13 CFG– + -------------------------------------------------- 1– 0.42 2 50 0– 50 0– 50 CHQ– + -------------------------------------------------- 1– += 0.43 2 25 0– 25 0– 25 CSN– + ------------------------------------------------- 1– 0.52 2 100 0– 100 0– 100 COC– + -------------------------------------------------------- 1– 0.20 2 100 0– 100 0– SSC 0– + ------------------------------------------------ 1– 0.10 2 100 0– 100 0– AOF 0– + ------------------------------------------------- 1– +0.20– RI 0.56 26 13 13 CFG– + -------------------------------------- 1– 0.42 100 50 50 CHQ– + -------------------------------------- 1– += 0.43 50 25 25 CSN– + ------------------------------------- 1– 0.52 200 100 100 COC– + -------------------------------------------- 1– 0.20 200 100 SSC 0– + ---------------------------------------- 1– 0.10 200 100 AOF 0– + ------------------------------------------ 1– – 0.20+ Table 5. Weighting factors of the indicator variables Indicator variables PC 1 CFG -0.560 SSC 0.204 CHQ -0.423 CSN -0.430 COC -0.520 AOF 0.103
- 8. 8 Bachtiar, I. et al. by coral communities. A transect with 100% algal or soft coral (AOF) covers will have a resilience index of 0.047, whileatransectthathas100%rublecoverwillhavearesilience index of 0.094. Regarding these situations, coral population recovery is still possible and dependent on other ecological processes. The present index can be used to compare the resilience status of coral reefs among locations and times. Fig. 3A shows coral reef resilience indices among four locations in theProvinceofKepulauanRiau,Indonesia.Usingthisindex we could also determine that coral reef resilience was significantly different among the four locations (F = 3.935, df = 3,225, P < 0.01). Coral reef resilience is dynamic as a results of its interaction with fluctuating stress and disturbances from the surrounding environment over time. Fig. 3B shows how the present index compares temporal variation in coral reef resilience. These figures confirmed the practical uses of the index to detect resilience variation spatially and temporally. 4. Discussion The present study shows that assessment of coral reef resilience level can be carried out using LIT data. Using this index,managersareabletomoreobjectivelycompareresilience levels spatially among coral reefs and more accurately make decisions on coral reef management. They could also describe Fig. 2. Performance of resilience index in relation to COC, AOF, and SSC Fig. 3. Comparison of coral reef resilience indices among locations (A) and times (B). Error bars indicate 1 SE
- 9. Practical Resilience Index for Coral Reef Assessment 9 temporal variation of the resilience level to predict future trends for management purposes. The present resilience index is merely ecological resilience. Severalauthorshaveproposedemployingamorecomplicated resilience measurement which covers not only ecological butalsophysicaloceanography,socialfactors,andmanagement practices(Maynardet al. 2010;Obura and Grimsditch2009; Cummingetal.2016).Suchcoralreefresiliencemeasurements are more comprehensive than the present study, but they are also impractical for coral reef managers. The present study provides a complementary resilience index to Maynard’s index. The Maynard’s index provides a broad assessment of the resilience state, while this present resilience index focuses on ecological assessment. The resilience state represented by the index may only operatemaximallyundergoodmanagementpractices.Improper management could impose additional stress and disturbances on coral reefs. The resilience index should be viewed as the resiliencestateofcoralreefecosystemsatthetimeofassessment. A resilience index will likely decrease in response to disturbances, due to reduction in coral cover, loss of coral functional groups, increasing coverage of rubble, or increasing algae and other fauna cover. The present index is very practical. A coral taxonomist or biologist is not always available in many provinces in Indonesia. Coralgenera richness (CGR) is thereforeomitted fromtheoperationalresilienceindicators.OburaandGrimsditch (2009) resilience assessment uses 61 indicators covering ecological, social, physical, and management factors. This complex resilience measurement would require a high level of financial support and a lot of expertise. On the other hand, the use of the presentindex can be utilized by nearly all coral reef managers with basic training experience in the LIT method and a spreadsheet computer program. The resilience index enable managers to make the right management decisions. When SSC is high due to coastal erosion, a possible management response is to deploy suitable settlement substrate to reduce SSC cover, for example concrete blocks (Bachtiar 2002) or reef balls (Bachtiar and Prayogo 2010).Howeverthiswouldonlybepracticaliferosionorthe disturbance could be reduced over the long term. The index may be used as an alternative and predictive measurementofcoralreefrecoveryafterdisturbance.Recovery of the ecosystem may be measured in terms of time to return to a former pre-disturbance resilience index. Conventional recovery time has mostly been measured as returning coral cover(Golbuuetal.2007;Gilmouretal.2013;Grahametal. 2015) and community composition or species diversity (Brown and Suharsono 1990; Done et al. 1991; Smith et al. 2008). Returningcoral covercould takeplaceover adecade, but returning species diversity might need many decades. The present index supports the efforts of Timpane-Padgham et al. (2017) to integrate resilience metrics in coral reef restorationefforts.Usingtheresilienceindex,coralreefrecovery doesnot only meanrecoveryof coralcoverbutalsofunctional group diversity, habitat quality and recruitment. Thepresentresilienceindexwouldbegloballyapplicable, sincereferencepointsofallvariableindicatorsweretheoretically their maximum values. Only the CSN value was determined from data collected from Indonesian waters. Since many Indonesian waters show the best places for coral growth (Tomascik et al.1996)andcenter of coraldiversity(Veron et al.2009),itmightbeinferredthatthebestinIndonesiamight also represent the best state worldwide. This means that the maximum value of CSN on the index would also be the maximum value in most other coral reef areas in the world. Furthermore, the index can be easily adapted to various transect lengths by using new CSN values related to the transect length, for example 30 or 50 meter length. The CSN variable is additively linear to transect length. The resilience index needs to be interpreted in relation to a specific disturbance. Carpenter et al. (2001) advised that resiliencemeasurementshouldclearlycomparetheresilienceof ‘what to what’. Interpretation of the resilience index used in this study should be developed for that purpose. When a coral reef with a resilience index of 0.450, for example, is reduced to 0.250 due to coral bleaching, the length of time for recovery to its former state is important. Time series data from permanently laid LIT, particularly in areas of mass coral mortality related to bleaching, will assist in further development and interpretation of the resilience index. It is recommended that the index should be validated through applicationbycoralreefresearchersfromotherIndo-Pacific regions. In the future, index interpretation should also be developed in relation to human-induced acute disturbances, such as blastfishingandshipgrounding.Indexinterpretationinrelation tochronicdisturbancemightbebestusedasawarningsignalof a potential greater ecological loss if it is accompanied by an acute disturbance. This system of resilience index assessment needs to be developed as a standard protocol.
- 10. 10 Bachtiar, I. et al. Acknowledgements This paper is supported by USAID through Sustainable Higher Education Research Alliances (SHERA) Program – Center for Collaborative Research Animal Biotechnology and CoralReefFisheries(CCRANBIOCORE).Authorswouldlike to thank Dr. Pamela Hallock-Muller of USF USA, Dr. Claudia BaldwinofUSCAustraliaandDr.AustinHumphriesofURIUSA fortheirassistancesonwritingand restructuringthemanuscript. References Babcock R, Davies P (1991) Effects of sedimentation on settlement of Acropora millepora. Coral Reefs 9:205–208 Bachtiar I (2002) Promoting recruitment of scieractinian corals using artificial substrate in the Gill Indah, Lombok Barat, Indonesia. In: Moosa MK, Soemodihardjo S, Soegiarto A, Romimohtarto K, Nontji A, Soekarno, Suharsono (eds) Proceedings 9th International Coral Reefs Symposium, Bali, 23–27 Oct 2000, pp 425–430 Bachtiar I,Prayogo W (2010) Coral recruitment on reef ball modules at the Benete Bay, Sumbawa island, Indonesia. J Coast Dev 13:119–125 Bahartan K, Zibdah M, Ahmed Y, Israel A, Bricker I, Abelson A (2010) Macroalgae in the coral reefs of Eilat (Gulf of Aqaba, Red Sea) as a possible indicator of reef degradation. Mar Pollut Bull 60:759–764 Bellwood DR, Hughes TP, Folke C, Nyström M (2004) Confronting the coral reef crisis. Nature 429:827–833 Bellwood DR, Hughes TP, Hoey AS (2006) Sleeping fractional group drives reef recovery. Curr Biol 16:2434–2439 Brown B, Suharsono (1990) Damage and recovery of coral reefs affected by El Nino related seawater warming in the Thousand Islands, Indonesia. Coral Reefs 8:163–170 Carpenter RC (1990) Mass mortality of Diadema antillarum: II. effectsonpopulationdensitiesandgrazingintensityofparrotfishes and surgeonfishes. Mar Biol 104:79–86 Carpenter C, Walker B, Anderies JM, Abel N (2001) From metaphor to measurement: resilience of what to what? Ecosystems 4:765–781 ChealAJ,McNeilMA,CrippsE,EmslieMJ,JonkerM,Schaffelke B, Sweatman H (2010) Coral-macroalgal phase shifts or reef resilience:linkswithdiversityandfunctionalrolesofherbivorous fishes on the Great Barrier Reef. Coral Reefs 29:1005–1015 ClarkeKR,SomerfieldPJ,GorleyRN(2008)Testingofnullhypotheses in exploratory community analyses: similarity profiles and biota-environment linkage. J Exp Mar Biol Ecol 366:56–69 Cumming GS, Morrison TH, Hughes TP (2016) New directions for understandingthespatialresilienceofsocial–ecologicalsystems. Ecosystems 20(4):649–664 Cruz ICS, Meira VH, de Kikuchi RKP, Creed JC (2016) The role of competition in the phase shift to dominance of the zoanthid Palythoacf. variabilisoncoral reefs.Mar EnvironRes 115:28–35 Done TJ (1982) Patterns in the distribution of coral communities across the Central Great Barrier Reef. Coral Reefs 1:95–107 Done TJ (1992) Phase shifts in coral reef communities and their ecological significance. Hydrobiologia 247:121–132 Done TJ, Dayton PK, Dayton AE, Steger R (1991) Regional and local variability in recovery of shallow coral communities: Moorea, French Polynesia and central Great Barrier Reef. Coral Reefs 9:183–192 English S, Wilkinson C, Baker V (1994) Survey manual for tropical marine resources. Australian Institute of Marine Sciences, Townsville, 368 p ErdingerEN,RiskMJ(2000)Reefclassificationbycoralmorphology predicts coral reef conservation value. Biol Conserv 92:1–13 Fox HE, Pet JS, Dahuri R, Caldwell RL (2003) Recovery in rubble fields: long-term impacts of blast fishing. Mar Pollut Bull 46:1024–1031 Fox RJ, Bellwood DR (2008) Remote video bioassays reveal the potential feeding impact of the rabbitfish Siganus canaliculatus (f: Siganidae) on an inner-shelf reef of the Great Barrier Reef. Coral Reefs 27:605–615 Gilmour JP, Smith LD, Heyward AJ, Baird AH, Pratchett MS (2013) Recovery of an isolated coral reef system following severe disturbance. Science 340:69–71 Golbuu Y, Victor S, Penland L, Idip Jr D, Emaurois C, Okaji K, Yukihira H, Iwase A, van Woesik R (2007) Palau’s coral reefs show differential habitat recovery following the 1998-bleaching event. Coral Reefs 26:319–332 Graham NAJ, Jennings S, MacNeil A, Mouillot D, Wilson SK (2015) Predicting climate-driven regime shifts versus rebound potential in coral reefs. Nature 518:94–97 Guest JR, Low J, Tun K, Wilson B, Ng C, Raingeard D, Ulstrup KE, Tanzil JTI, Todd PA, Toh PA, McDaugald D, Chou LM, Steinberg PD (2016) Coral community response to bleaching onahighly disturbed reef. SciRep-UK15(6):385.doi:10.1038/srep20717 Hodgson G (1990) Sediment and the settlement of larvae of the reef coral Pocillopora damicornis. Coral Reefs 9:41–43 Hoegh-GuldbergO,MumbyPJ,HootenAJ,SteneckRS,Greenfield P, Gomez E, Harvell CD, Sale PF, Edwards AJ, Caldeira K, Knowlton N, Eakin CM, Iglesias-Prieto R, Muthiga N, Bradbury RH, Dubi A, Hatziolos ME (2007) Coral reefs under rapid climate change and ocean acidification. Science 318:1737–1742 Hoey AS, Bellwood DR (2008) Cross-shelf variation in the role of parrot fishes on the Great Barrier Reef. Coral Reefs 27:37–47 Holling CS (1973) Resilience and stability of ecological systems. Annu Rev Ecol Syst 4:1–23 Hughes TP (1994) Catastrophes, phase shifts and large-scale degradation of Caribbean coral reef. Science 65:1547–1551 Hughes TP, Rodrigues MJ, Bellwood DR, Ceccarelli D, Hoegh- Guldberg O, McCook L, Moltschaniwsky N, Pratchett MS, Steneck RS, Willis B (2007) Phase shifts, herbivory, and the
- 11. Practical Resilience Index for Coral Reef Assessment 11 resilience of coral reefs to climate change. Curr Biol 17:1–6 Jackson JBC (1997) Reefs since Columbus. Coral Reefs 16(Suppl): S23–S32 Jackson JBC, Kirby MX, Berger WH, Bjomdal KA, Botsford LW, Bourque BJ, Bradbury RH, Cooke R, Erlandson J, Estes JA, Hughes TP, Kidwell S, Lange CB, Lenihan HS, Pandolfi JM, Peterson CH, Stenneck RS, Tegner MJ, Warner RR (2001) Historical overfishing and the recent collapse of coastal ecosystems. Science 293:629–637 Kleypas JA, Buddemeier RW, Archer D, Gattuso J-P, Langdon C, Opdyke BN (1999) Geochemical consequences of increased atmospheric carbon dioxide on coral reefs. Science284:118–120 Lam VYY, Doropoulus C, Mumby PJ (2017) The influence of resilience-based management on coral reef monitoring: a systematic review. PLoS One 12(2):e0172064. doi:10.1371/ journal.pone.0172064 Ledlie MH, Graham NAJ, Bythell JC, Wilson SK, Jennings S, Polunin NVC, Hardcastle J (2007) Phase shifts and the role of herbivoryinthe resilienceofcoral reefs. CoralReefs26:641–653 Littler MM, Littler DS, Brooks BL (2006) Harmful algae on tropical coral reefs: bottom-up eutrophication and top-down herbivory. Harmful Algae 5:565–585 Lundberg J, Moberg F (2003) Mobile link organisms and ecosystem functioning:implicationsforecosystemresilienceandmanagement. Ecosystems 6:87–98 Marsh LM, Bradbury RH, Reichelt RE (1984) Determination of the physical parameters of coral distributions using line transect data. Coral Reefs 2:175–180 Maynard JA, Marshall PA, Johnson JE, Harman S (2010) Building resilienceintopracticalconservation:identifyinglocalmanagement responses to global climate change in the southern Great Barrier Reef. Coral Reefs 29:381–391 McClanahan TR, Donner SD, Maynard JA, MacNeil MA, Graham NAJ, Maina J, Baker AC, Alemu I JB, Beger M, Campbell SJ, Darling ES, Eakin CM, Heron SF, Jupiter SD, Lundquist CJ, McLeod E, Mumby PJ, Paddack MJ, Selig ER, van Woesik R (2012) Prioritizing key resilience indicators to support coral reef management in a changing climate. PLoS One 7:e42884. doi:10.1371/journal.pone.0042884 Miller K, Mundy C (2003) Rapid settlement in broadcast spawning corals: implications for larval dispersal. Coral Reefs 22:99–106 Mora C, Graham N J, Nystrom M (2016) Ecological limitations to the resilience of coral reefs. Coral Reefs 35(4):1271–1280 Ninio R, Meekan MG (2002) Spatial patterns in benthic communities and the dynamics of a mosaic ecosystem on the Great Barrier Reef, Australia. Coral Reefs 21:95–103 NystromM,GrahamAJ,LokrantzJ,NorströmAV (2008)Capturing the cornerstones of coral reef resilience: linking theory to practice. Coral Reefs 27:795–809 Obura DO, Grimsditch G (2009) Resilience assessment of coral reefs - assessment protocol for coral reefs, focusing on coral bleaching and thermal stress. IUCN, Gland, 70 p Orwin KH, Wardle DA (2004) New indices for quantifying the resistance and resilience of soil biota to exogenous disturbances. Soil Biol Biochem 36:1907–1912 Peterson G, Allen CR, Holling CS (1998) Ecological resilience, biodiversity, and scale. Ecosystems 1:6–18 Primpas I, Tsirtsis G, Karydis M, Kokkoris GD (2010) Principal component analysis: development of a multivariate index for assessing eutrophication according to the European water framework directive. Ecol Indic 10:178–183 Roberts CM, Ormond RFG (1987) Habitat complexity and coral reef fish diversity and abundance on Red Sea fringing reefs. Mar Ecol-Prog Ser 41:1–8 Roff G, Mumby PJ (2012) Global disparity in the resilience of coral reefs. Trends Ecol Evol 27:404–413 Rogers CS, Gilnack M, Fitz III HC (1983) Monitoring of coral reefs with linear transects: a study of storm damage. J Exp Mar Biol Ecol 66:285–300 Smith LD, Gilmour JP, Heyward AJ (2008) Resilience of coral communitiesonanisolatedsistemofreefsfollowingcatastrophic mass-bleaching. Coral Reefs 27:197–205 StargerCJ,BarberPH,Ambariyanto,BakerAC(2010)Therecovery of coral genetic diversity in the Sunda Strait following the 1883 eruption of Krakatau. Coral Reefs 29:547–565 Suharsono (2008) Managing Indonesian coral reefs: lessons from coralreefrehabilitationandmanagementprogram.In:Proceedings 11th International Coral Reef Symposium, Florida, 7–11 Jul 2008, pp 1159–1161 Timpane-Padgham BL, Beechie T, Klinger T (2017) A systematic review of ecological attributes that confer resilience to climate changeinenvironmentalrestoration.PLoSOne12(3):e0173812. https://doi.org/10.1371/journal.pone.0173812 Tkachenko KS, Wu BJ, Fang LS, Fan TY (2007) Dynamics of a coral reefcommunity aftermass mortality of branchingAcropora corals and an outbreak of anemones. Mar Biol 151:185–194 Tomascik T, van Woesik R, Mah AJ (1996) Rapid coral colonization of a recent lava flow following a volcanic eruption, Banda Islands, Indonesia. Coral Reefs 15:169–175 Van Moorsel GWNM (1985) Disturbance and growth of juvenile corals (Agaricia humilis and Agaricia agaricites, Scleractinia) in natural habitats on the reef of Curacao. Mar Ecol-Prog Ser 24:99–112 Veron JEN, Devantier LM, Turak E, Green AL, Kininmonth S, Stafford-Smith M, Peterson N (2009) Delineating the coral triangle. Galaxea 11:91–100 Williams DE, Miller MW, Kramer KL (2008) Recruitment failure in Florida Keys Acropora palmata, a threatened Caribbean coral. Coral Reefs 27:697–705 Wilson SK, Graham NAJ, Polunin NVC (2007) Appraisal of visual assessments of habitat complexity and benthic composition on coral reefs. Mar Biol 151:1069–1076 Publisher’sNote SpringerNatureremainsneutralwithregardtojurisdictional claims in published mapsand institutionalaffiliations